ashwin/nxtgauge-backend-rust
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

fix(ai): use Ollama cluster URL and gemma3:270m model defaults

Browse source 
- Default OLLAMA_BASE_URL to http://ollama.nxtgauge-ai.svc.cluster.local:11434
- Default OLLAMA_CHAT_MODEL to gemma3:270m (matches gitops configmap)
This commit is contained in:
Tracewebstudio Dev 2026-04-15 19:54:58 +02:00
parent 430711a0ae
commit ebc0a29437
1 changed files with 6 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

12
apps/users/src/handlers/ai.rs
View file

@ -13,7 +13,7 @@ pub fn ai_router() -> Router<AppState> {
Router::new() Router::new()
.route("/chat/message", post(ai_chat_message)) .route("/chat/message", post(ai_chat_message))
.route("/tickets/create", post(ai_create_ticket)) .route("/tickets/create", post(ai_create_ticket))
.route("/tickets/:id", get(ai_get_ticket)) .route("/tickets/{id}", get(ai_get_ticket))
.route("/forms/extract", post(ai_extract_form)) .route("/forms/extract", post(ai_extract_form))
} }
@ -46,7 +46,7 @@ struct OllamaGenerateResponse {
} }
async fn call_ollama(state: &AppState, model: &str, prompt: &str) -> Result<String, String> { async fn call_ollama(state: &AppState, model: &str, prompt: &str) -> Result<String, String> {
let base_url = std::env::var("OLLAMA_BASE_URL").unwrap_or_else(|_| "http://localhost:11434".to_string()); let base_url = std::env::var("OLLAMA_BASE_URL").unwrap_or_else(|_| "http://ollama.nxtgauge-ai.svc.cluster.local:11434".to_string());
let url = format!("{}/api/generate", base_url); let url = format!("{}/api/generate", base_url);
let req = OllamaGenerateRequest { let req = OllamaGenerateRequest {
@ -130,8 +130,8 @@ async fn ai_chat_message(
State(state): State<AppState>, State(state): State<AppState>,
Json(body): Json<OllamaChatRequest>, Json(body): Json<OllamaChatRequest>,
) -> impl IntoResponse { ) -> impl IntoResponse {
let ollama_base = std::env::var("OLLAMA_BASE_URL").unwrap_or_else(|_| "http://localhost:11434".to_string()); let ollama_base = std::env::var("OLLAMA_BASE_URL").unwrap_or_else(|_| "http://ollama.nxtgauge-ai.svc.cluster.local:11434".to_string());
let model = std::env::var("OLLAMA_CHAT_MODEL").unwrap_or_else(|_| "smollm2:360m".to_string()); let model = std::env::var("OLLAMA_CHAT_MODEL").unwrap_or_else(|_| "gemma3:270m".to_string());
let default_conversation = Uuid::new_v4().to_string(); let default_conversation = Uuid::new_v4().to_string();
let conversation_id = body.conversation_id.unwrap_or_else(|| default_conversation); let conversation_id = body.conversation_id.unwrap_or_else(|| default_conversation);
@ -291,8 +291,8 @@ async fn ai_extract_form(
State(state): State<AppState>, State(state): State<AppState>,
Json(body): Json<FormExtractBody>, Json(body): Json<FormExtractBody>,
) -> impl IntoResponse { ) -> impl IntoResponse {
let ollama_base = std::env::var("OLLAMA_BASE_URL").unwrap_or_else(|_| "http://localhost:11434".to_string()); let ollama_base = std::env::var("OLLAMA_BASE_URL").unwrap_or_else(|_| "http://ollama.nxtgauge-ai.svc.cluster.local:11434".to_string());
let model = std::env::var("OLLAMA_CHAT_MODEL").unwrap_or_else(|_| "smollm2:360m".to_string()); let model = std::env::var("OLLAMA_CHAT_MODEL").unwrap_or_else(|_| "gemma3:270m".to_string());
let form_type = body.form_type.unwrap_or_else(|| "generic".to_string()); let form_type = body.form_type.unwrap_or_else(|| "generic".to_string());